Unlock instant, AI-driven research and patent intelligence for your innovation.

Face attribute migration method based on image segmentation and generative adversarial network

A technology of attribute migration and image segmentation, which is applied in the field of image processing, can solve the problems of not being able to distinguish between real pictures and generated pictures, ignoring the unity of style, etc., and achieve the effect of reducing fragmentation and realistic face images

Pending Publication Date: 2021-12-28
NORTHWESTERN POLYTECHNICAL UNIV
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In this way, G and D constitute a dynamic "game process". After non-stop games, the G network and D network have reached the Nash equilibrium state, and it is no longer possible to distinguish between real pictures and generated pictures.
In the current GAN-based face attribute migration, the unified style after attribute migration is often ignored

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Face attribute migration method based on image segmentation and generative adversarial network
  • Face attribute migration method based on image segmentation and generative adversarial network
  • Face attribute migration method based on image segmentation and generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The method of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments of the present invention.

[0032] It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and examples.

[0033] It should be noted that the terminology used here is only for describing specific implementations, and is not intended to limit the exemplary implementations according to the present application. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural, and it should also be understood that when the terms "comprising" and / or "comprising" are used in this specification, they mean There are features, steps, operations, means, components and / or combinations thereof....

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a face attribute migration method based on image segmentation and a generative adversarial network, and the method comprises the following steps: setting a basic network framework of face attribute migration, and setting basic parameters of the network; using the CelebA data set to train the U-Net face image segmentation model; segmenting a face image by using the trained U-Net network, segmenting a face attribute needing to be migrated from a source image, and fusing the face attribute to a target face image; inputting the target image and the fused image into a generative adversarial network to enable the fused image to accord with the style of the target image; and recognizing a face region by using a cascade classifier based on Haar features, and carrying out face attribute migration. According to the face attribute migration method provided by the invention, various face attributes can be migrated to the target face image, the migrated and replaced part is smoother through generative adversarial learning, the split feeling is reduced, and a more vivid face image can be obtained.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a face attribute migration method based on image segmentation and generating an adversarial network. Background technique [0002] Deep forgery technology is a combination of deep learning and fake (fake). Taking human faces as an example, it can perform attribute transformation and style transfer on the image of the target person's face in order to confuse the audience. Deep forgery has the characteristics of high simulation and strong deception, and can be applied to tasks such as sound synthesis, video resolution restoration, and image art style transfer. [0003] Face-swapping forgery refers to replacing a known face with a target face. At present, the main face-swapping tools include open source FaceSwap, Deep-FaceSwap, Faceswap-GAN, DeepFaceLab, etc. Attribute tampering achieves the purpose of tampering by modifying or adding or subtracting some attrib...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/00G06K9/34G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/2148
Inventor 耿杰邓号蒋雯邓鑫洋
Owner NORTHWESTERN POLYTECHNICAL UNIV